Hugging Face's logo
David Faragó's picture
27 6 2

David Faragó

dball
zaursamedov1's profile picture catastropiyush's profile picture
·

AI & ML interests

None yet

Organizations

mediform.io's profile picture

dball's activity

commented a paper 22 days ago
commented a paper about 1 month ago
New activity in PatronusAI/Llama-3-Patronus-Lynx-8B-Instruct-v1.1 about 2 months ago

Error when deploying an inference endpoint

2
#1 opened about 2 months ago by dball
commented 4 papers 6 months ago
commented 2 papers 7 months ago
New activity in dball/zephyr-7b-sft-qlora 8 months ago

Is the drop in many metrics expected? Why do SFT first if it makes the model worse? Why not do DPO directly on the mistral model?

1
#1 opened 8 months ago by dball
New activity in dball/zephyr-7b-dpo-qlora 8 months ago

Is the high drop in GSM8K usual?

1
#1 opened 8 months ago by dball
New activity in teknium/CollectiveCognition-v1.1-Mistral-7B 9 months ago

Training code?

4
#5 opened 12 months ago by umarmnaq
New activity in malhajar/Mistral-7B-v0.2-meditron-turkish 9 months ago

Training script?

1
#1 opened 9 months ago by dball
New activity in fxmarty/tiny-llama-fast-tokenizer about 1 year ago

Bugs when loading tokenizer or model

1
#3 opened about 1 year ago by dball
New activity in Yhyu13/oasst-rlhf-2-llama-30b-7k-steps-hf about 1 year ago

How did you merge?

1
#1 opened about 1 year ago by dball
New activity in garage-bAInd/GPlatty-30B about 1 year ago

Curious about model design

2
#2 opened about 1 year ago by surya-narayanan